JP3970655B2 - Control device for internal combustion engine - Google Patents

Description

【００３６】
図２は、この発明に従う制御装置の機能ブロック図である。各機能ブロックで表される機能は、典型的にはコンピュータプログラムによって実行される。代替的には、各機能ブロックで表される機能を実行するよう構成された任意のハードウェアによって、各機能ブロックを実現してもよい。
【００３７】
リーク判定部８０は、所定の条件が満たされたとき、パージ系３１にリークがあるかどうかを判定する。所定の条件には、１）パージ処理が実行中である、２）エンジンの運転状態が所定の定常状態にある、３）車速の変化が小さい、４）空燃比補正係数が所定値範囲にある、等を含めることができる。代替的に、他の条件を用いてもよいし、異なる条件を追加してもよい。リーク判定部８０は、所定の条件が満たされたとき、リーク判定の実行を許可するために、フラグＦ＿ＥＶＰＬＫＭに１をセットする。
【００３８】
リーク判定部８０は、フラグＦ＿ＥＶＰＬＫＭに１がセットされたならば、以下の手順に従い、パージ系３１のタンク系およびキャニスタ系のリーク判定を実行する。
【００３９】
［タンク系のリーク判定］
１）バイパス弁２４およびベントシャット弁２６を開き、パージ制御弁３０を閉じて、燃料タンク９を大気に開放する。
【００４０】
２）バイパス弁２４およびパージ制御弁３０を閉じて、大気圧から正圧に上昇する単位時間あたりの圧力変動量を補正値として測定する補正チェックモードを実施する。
【００４１】
３）バイパス弁２４を開き、ベントシャット弁２６を閉じ、パージ制御弁３０を制御しながら、タンク系を所定の圧力にまで安定的に減圧する減圧モードを実施する。
【００４２】
４）タンク系が所定の負圧状態になった後、すべての弁２４、２６および３０を閉じ、復帰する単位時間当たりの圧力変動量を測定するリークチェックモードを実施する。
【００４３】
５）リークチェックモード中の単位時間あたりの圧力変動量から、補正チェックモード中の単位時間あたりの圧力変動量に所定の係数を掛けた値を減算する。その結果算出された値に基づいて、タンク系のリークを判定する。該算出された値が小さければ、タンク系の負圧状態がほぼ保持されており、タンク系にはリークが無いと判定される。該算出された値が大きければ、タンク系にリークがあると判定される。
【００４４】
［キャニスタ系］
１）バイパス弁２４およびベントシャット弁２６を開き、パージ制御弁３０を閉じて、キャニスタ系を大気に開放する。
【００４５】
２）ベントシャット弁２６を閉じ、パージ制御弁３０を開いて、キャニスタ系を所定の圧力にまで安定的に減圧する。
【００４６】
３）キャニスタ系が所定の負圧状態になった後、すべての弁２４、２６および３０を閉じ、復帰する圧力変動量を測定する。変動量が小さければ、キャニスタ系の負圧状態がほぼ保持されており、リークが無いと判定される。変動量が大きければ、キャニスタ系にリークがあると判定される。
【００４７】
タンク系およびキャニスタ系のいずれか一方のリーク判定が実施される場合にも、当然ながら本発明を適用することができる。また、対象となる系を負圧にしてリーク判定を実施する手法として、上記以外の他の手法を用いてもよい。
【００４８】
機能ブロック８１〜８９に示される機能は、パージされる蒸発燃料の量に応じて燃料噴射量を補正する技術を実現する。
【００４９】
吸入空気量算出部８１は、式（４）に従い、吸入空気量ＱＡＩＲを算出する。
【００５０】
【数４】
ＱＡＩＲ＝ＴＩＭ×ＮＥ×２×ＫＱＡＩＲ×ＫＰＡ （４）
【００５１】
ここで、ＴＩＭは、前述した基本燃料噴射量を示す。ＫＱＡＩＲは、燃料噴射量を空気の流量に換算するための係数であり、固定値（たとえば、０．４５L/ms）を持つ。ＫＰＡは、吸気管圧力ＰＢＡに応じた流量の変動を補正するための係数である。
【００５２】
吸入空気量算出部８１は、さらに、吸入空気量ＱＡＩＲに対する蒸発燃料の割合、すなわち基本パージ量ＱＰＧＣＢＡＳＥを、式（５）に従って算出する。
【００５３】
【数５】
ＱＰＧＣＢＡＳＥ＝ＱＡＩＲ×ＫＱＰＧＢ （５）
【００５４】
ここで、ＫＱＰＧＢは目標パージ率を示しており、たとえば０．０４である。この場合、吸入空気量ＱＡＩＲの４％に蒸発燃料を含めることになる。
【００５５】
パージ流量算出部８２は、基本パージ量ＱＰＧＣＢＡＳＥに基づいて、目標パージ流量ＱＰＧＣＭＤを式（６）に従って算出する。目標パージ流量ＱＰＧＣＭＤは、今回のサイクルにおいてエンジンの吸気系にパージするパージ流量の目標値を表す。
【００５６】
【数６】
ＱＰＧＣＭＤ＝ＱＰＧＣＢＡＳＥ×ＫＰＧＴ （６）
【００５７】
ここで、ＫＰＧＴはパージ流量係数であり、１以下の値を持つ。係数ＫＰＧＴを用いて、目標パージ流量ＱＰＧＣＭＤの量を制御することができる。係数ＫＰＧＴは、運転状態に応じて算出される。
【００５８】
さらに、パージ流量算出部８２は、吸入空気量ＱＡＩＲに基づいて、今回のサイクルでパージするパージ流量ＱＰＧＣを式（７）に従って算出する。
【００５９】
【数７】
ＱＰＧＣ(ｋ)＝ＱＰＧＣ(ｋ−１)＋（ＱＡＩＲ×ＫＤＱＰＧＣ） （７）
【００６０】
ここで、ｋはサイクルを識別する数字であり、（ｋ）は今回のサイクルを示し、（ｋ−１）は前回のサイクルを示す。ＫＤＱＰＧＣは予め決められた固定値（たとえば、０．００３）である。式（７）から明らかなように、パージ流量ＱＰＧＣは、徐々に目標パージ流量ＱＰＧＣＭＤに達するように制御される。
【００６１】
デューティ算出部８３は、パージ流量算出部８２から受け取ったパージ流量ＱＰＧＣがパージされるように、パージ制御弁を駆動するデューティ比ＰＧＣＭＤを、式（８）に従って算出する。デューティ比は、パージ制御弁が開弁する比率を表す。
【００６２】
【数８】

【００６３】
ＫＤＵＴＹは、パージ流量をデューティ比に換算するための係数であり、固定値（たとえば、３．８%・min/L）を持つ。差圧に応じてパージ制御弁の開度が変化するので、ＫＤＰＢＧは、それを補正するための係数である。ＰＧＣＭＤ０は、パージ流量ＱＰＧＣに対応するデューティ比を表しており、以下目標デューティ比と呼ぶ。バッテリ電圧ＶＢおよび吸気管圧力ＰＢＡに依存してパージ制御弁が開き始めるまでに何らかの遅れが生ずるので、ＤＰＧＣＶＢＸおよびＤＰＧＣ０は、それぞれ、この遅れ（以下、無効時間と呼ぶ）を補正する係数である。
【００６４】
デューティ算出部８３は、デューティ比ＰＧＣＭＤに対して、所定の上限値および下限値でリミット処理を行い、最終デューティ比ＤＯＵＴＰＧＣを出力する。こうして、パージ制御弁の開度は、最終デューティ比ＤＯＵＴＰＧＣに従って制御される。
【００６５】
輸送遅れ算出部８４は、エンジン回転数ＮＥに基づいて、パージ輸送遅れＣＰＧＤＬＹＲＸを算出する。パージ輸送遅れＣＰＧＤＬＹＲＸは、蒸発燃料がパージ通路にパージされてからエンジンの吸気系に送られるまでの時間的な遅れを示す。パージ輸送遅れＣＰＧＤＬＹＲＸは整数ｎで表され、ｎが大きくなるにつれ輸送遅れが大きいことを示す。代替的に、パージ輸送遅れを、エンジン回転数ＮＥの代わりに吸入空気量ＱＡＩＲに基づいて算出するようにしてもよい。
【００６６】
空燃比制御部８５は、ＬＡＦセンサ３２からの出力に基づいて、内燃機関に供給される混合気の空燃比をフィードバック制御によって目標空燃比に収束させるための空燃比係数ＫＡＦを算出する。
【００６７】
空燃比学習値算出部８６は、空燃比制御部８５によって算出された空燃比係数ＫＡＦに基づいて、パージ有り学習値ＫＲＥＦと、パージ無し学習値ＫＲＥＦＸとを、以下の式（９）に基づいて算出する。前述したように、（ｋ）は今回のサイクルを示し、（ｋ−１）は前回のサイクルを示す。
【００６８】
【数９】

【００６９】
ここで、＃ＣＲＥＦおよび＃ＣＲＥＦＸは、それぞれ、重み付け係数を表す所定値である。この値で、前回のサイクルで算出されたＫＲＥＦ(k-1)およびＫＲＥＦＸ(k-1)に対し、どれだけ空燃比補正係数ＫＡＦを反映させるかを決定する。
【００７０】
ベーパ濃度係数算出部８７は、空燃比制御部８５によって算出された空燃比補正係数ＫＡＦ、空燃比学習値算出部８６によって算出された学習値ＫＲＥＦＸおよびＫＲＥＦに基づいて、ベーパ濃度係数ＫＡＦＥＶを算出する。
【００７１】
目標パージ補正係数算出部８８は、以下の式（１０）に従って、目標パージ補正係数ＫＡＦＥＶＡＣＺを算出する。
【００７２】
【数１０】

【００７３】
ＰＧＲＡＴＥおよびＱＲＡＴＥは、式（１１）および（１２）で表される。
【００７４】
【数１１】

【数１２】
ＱＲＡＴＥ＝ＱＰＧＣ／ＱＰＧＣＢＡＳＥ （１２）
【００７５】
「ＰＧＲＡＴＥ×ＱＲＡＴＥ」は、今回のサイクルにおいてエンジンの吸気系にパージされるパージ流量の割合を示す。具体的に説明すると、前述したようにＤＯＵＴＰＧＣは最終デューティ比を表しているので、式（１１）に示されるデューティレートＰＧＲＡＴＥは、パージ流量ＱＰＧＣの目標デューティ比ＰＧＣＭＤ０に対する、実際のデューティ比（無効時間を差し引いたもの）の割合を示す。したがって、「ＰＧＲＡＴＥ×ＱＰＧＣ」は、今回のサイクルにおいてパージ制御弁によってパージされる実際のパージ流量を示す。
【００７６】
一方、基本パージ流量ＱＰＧＣＢＡＳＥは、式（５）を参照して前述したように、今回のサイクルにおける吸入空気量ＱＡＩＲに含めることのできる蒸発燃料の流量を示す。燃料はベーパ濃度係数ＫＡＦＥＶに依存して表されるので、式（１０）に示されるように、「ＰＧＲＡＴＥ×ＱＲＡＴＥ」にベーパ濃度係数ＫＡＦＥＶを乗算することにより、目標パージ補正係数ＫＡＦＥＶＡＣＺを求めることができる。
【００７７】
目標パージ補正係数算出部８８は、０〜（ｎ−１）の番号がそれぞれ付与された複数のバッファを有しており、今回のサイクルで算出された目標パージ補正係数をゼロ番目のバッファに格納し、前回のサイクルで算出された目標パージ補正係数を１番目のバッファに格納し、．．．というように、目標パージ補正係数を時系列に格納する。目標パージ補正係数算出部８８は、輸送遅れ算出部８４から蒸発燃料の輸送遅れＣＰＧＤＬＹＲＸの値ｎを受け取り、該ｎに対応する番号のバッファから、目標パージ補正係数ＫＡＦＥＶＡＣＺを抽出する。たとえば、受け取ったＣＰＧＤＬＹＲＸの値ｎが３ならば、３番目のバッファから目標パージ補正係数ＫＡＦＥＶＡＣＺを抽出する。
【００７８】
パージ補正係数算出部８９は、算出された目標パージ補正係数ＫＡＦＥＶＡＣＺに基づいて、要求燃料に対してパージ流量ＱＰＧＣが寄与する燃料量の割合を示すパージ補正係数ＫＡＦＥＶＡＣＴ（式（２）を参照して前述した）を算出する。
【００７９】
パージ補正係数算出部８９は、リーク判定部８０によってリーク判定が実施されていないとき（すなわち、フラグＦ＿ＥＶＰＬＫＭがゼロのとき）、式（１３）および（１４）に従って、パージ補正係数ＫＡＦＥＶＡＣＴを算出する。
【００８０】
【数１３】

【数１４】

【００８１】
ここで、ＫＫＥＶＧは高濃度補正係数を示しており、蒸発燃料の濃度が非常に高いと空燃比への影響が大きいので、これを補正するための係数である。高濃度補正係数ＫＫＥＶＧは、蒸発燃料の濃度が非常に高いときに、それに応じて１より大きい値が設定される。
【００８２】
漸増係数ＫＥＶＡＣＴは、所定の初期値（たとえば、ゼロ）から１．０まで徐々に増やされる係数である。漸増係数ＫＥＶＡＣＴは、パージが開始された時は初期値に設定され、その後、それぞれのサイクルで所定値だけ増やされていく。こうして、要求燃料量に寄与するパージ補正量を徐々に増やすことで、パージ処理を開始した時の空燃比の変動を防ぐ。パージカット中は、漸増係数ＫＥＶＡＣＴはゼロに設定される。
【００８３】
＃ＣＫＡＦＥＶＡＣは、重み付け係数を表す所定値である。この値で、前回のサイクルで算出されたパージ補正係数ＫＡＦＥＶＡＣＴ(k-1)に対し、どれだけ暫定パージ補正係数ＫＡＦＥＶＡＣを反映させるかを決定する。
【００８４】
パージ補正係数算出部８９は、リーク判定部８０によってリーク判定が実施されているとき（すなわち、フラグＦ＿ＥＶＰＬＫＭが１のとき）、式（１５）に従って、パージ補正係数ＫＡＦＥＶＡＣＴを算出する。
【００８５】
【数１５】
ＫＡＦＥＶＡＣＴ＝ＫＡＦＥＶＡＣＺ×ＫＫＥＶＧ 式（１５）
【００８６】
式（１４）と比較して明らかなように、リーク判定が実施されているときは、パージ補正係数ＫＡＦＥＶＡＣＴの算出に漸増係数ＫＥＶＡＣＴを使用しない。
【００８７】
パージ開始時には、漸増係数ＫＥＶＡＣＴで、要求燃料に対するパージ補正量の割合を絞ることにより、空燃比の変動を抑制することができる。しかしながら、前述したように、リーク判定中はパージ系を速やかに負圧にするため、パージ制御弁は強制的に所定の開度に開かれる。その結果、算出されたパージ補正量より多くの蒸発燃料がパージされ、空燃比がリッチ化し、運転状態を悪化させるおそれがある。これを回避するため、リーク判定中は、漸増係数ＫＥＶＡＣＴによるパージ補正量の漸増制御を実施しないようにする。こうして、リーク判定中は、パージ補正量として、算出された蒸発燃料の量が最初から設定されるので、空燃比が適正に維持される。
【００８８】
さらに、パージ補正係数算出部８９は、オープンループ制御においては、上記の式（１３）または（１５）に基づいて算出された補正係数ＫＡＦＥＶＡＣＴを、以下の式（１６）に従って補正する。
【００８９】
【数１６】

【００９０】
ここで、パージ有り学習値ＫＲＥＦは、パージが実行されている間に空燃比補正係数ＫＡＦをなまし平均した値である。パージ無し学習値ＫＲＥＦＸは、パージカットを実施している間の空燃比係数ＫＡＦをなまし平均した値である。式（１６）で示されるように、補正係数ＫＡＦＥＶＡＣＴは、パージ有り学習値とパージ無し学習値の偏差（ＫＲＥＦ−ＫＲＥＦＸ）で補正される。学習値ＫＲＥＦおよびＫＲＥＦＸは、直近に実行されたフィードバック制御時に算出されたものを用いる。
【００９１】
ベーパ濃度係数ＫＡＦＥＶは推定により算出されるので、運転状態によっては誤差を含むことがある。その結果、パージ補正係数ＫＡＦＥＶＡＣＴに“ずれ”が生じる。空燃比フィードバック制御が実行されている間は、空燃比補正係数ＫＡＦにより空燃比が目標空燃比に達するよう制御されるので、該ずれは修正されることができる。
【００９２】
しかしながら、オープンループ制御が実行されている間は、空燃比補正係数ＫＡＦが、たとえば値１．０に固定されるので、該ずれを空燃比補正係数ＫＡＦで修正することができない。したがって、オープンループ制御時は、フィードバック制御時に空燃比補正係数ＫＡＦによって修正されていたずれを求め、該ずれに応じた量で、パージ補正係数ＫＡＦＥＶＡＣＴを補正する。パージ有り学習値ＫＲＥＦとパージ無し学習値ＫＲＥＦＸの差は、この“ずれ”を表している。こうして、フィードバック制御時に含まれていたパージ補正係数ＫＡＦＥＶＡＣＴのずれが解消され、オープンループ制御においても、要求燃料に適合した量の燃料が内燃機関に供給される。
【００９３】
供給燃料算出部９０は、パージ補正係数算出部８９からパージ補正係数ＫＡＦＥＶＡＣＴを受け取り、上記の式（３）に従って燃料噴射量ＴＣＹＬを算出する。算出された燃料噴射量ＴＣＹＬは、燃料噴射弁を介してエンジンに供給される。こうして、パージ流量ＱＰＧＣが寄与する蒸発燃料量（すなわち、パージ補正量）を差し引いた量の燃料が、燃料噴射弁を介してエンジンに供給される。
【００９４】
図３は、吸入空気量ＱＡＩＲおよび基本パージ流量ＱＰＧＢＡＳＥを求めるフローチャートである。このフローは、たとえばＴＤＣセンサからＴＤＣパルス信号が出力されるたびに実行される。ステップＳ１０１において、前述した式（４）に従って吸入空気量ＱＡＩＲを求める。ステップＳ１０２において、前述した式（５）に従って基本パージ流量ＱＰＧＣＢＡＳＥを求める。
【００９５】
図４は、パージ制御弁を駆動する最終デューティ比ＤＯＵＴＰＧＣを求めるフローチャートである。このフローは、一定の時間間隔（たとえば８０ミリ秒）で繰り返し実行される。
【００９６】
ステップＳ１１１において、パージ流量ＱＰＧＣを求めるルーチン（図５）を実行する。ステップＳ１１２において、ステップＳ１１１で算出されたパージ流量ＱＰＧＣがゼロならば、蒸発燃料をパージしないことを示すので、デューティ比ＰＧＣＭＤにゼロを設定し（Ｓ１１３）、ステップＳ１２０に進む。ステップＳ１１２においてパージ流量ＱＰＧＣがゼロでなければ、ステップＳ１１４に進み、パージ制御弁を駆動する周期（たとえば、８０ミリ秒）を設定する。
【００９７】
ステップＳ１１５〜Ｓ１１９は、前述した式（８）に従ってデューティ比ＰＧＣＭＤを求める処理である。ステップＳ１１５において、バッテリ電圧に応じたパージ制御弁の無効時間を補正するため、ＤＰＧＣＶＢＸテーブルをアクセスし、バッテリ電圧ＶＢに基づいて無効時間ＤＰＧＣＶＢＸを求める。ＤＰＧＣＶＢＸテーブルは、バッテリ電圧が大きくなるほど、無効時間ＤＰＧＣＶＢＸが小さくなるよう設定されている。
【００９８】
ステップＳ１１６に進み、差圧に起因するパージ制御弁のデューティの変動を補正するため、ＫＤＰＢＧテーブルをアクセスし、吸気管圧力ＰＢＡに基づいて差圧補正値ＫＤＰＢＧを求める。ＫＤＰＢＧテーブルは、エンジン負荷が低くなるほど、差圧補正値ＫＤＰＢＧが大きくなるよう設定されている。ステップＳ１１７に進み、吸気管圧力ＰＢＡに応じたパージ制御弁の無効時間を補正するため、ＤＰＧＣ０テーブルをアクセスし、吸気管圧力ＰＢＡに基づいて無効時間ＤＰＧＣ０を求める。ＤＰＧＣ０テーブルは、負荷が大きくなるほど、無効時間ＤＰＧＣ０が大きくなるよう設定されている。ステップＳ１１８およびステップＳ１１９で、前述の式（８）に従い、デューティ比ＰＧＣＭＤを求める。
【００９９】
ステップＳ１２０〜Ｓ１２５は、デューティ比ＰＧＣＭＤに対するリミット処理を示す。ステップＳ１２０において、デューティ比ＰＧＣＭＤが予め決められた上限値ＤＯＵＴＰＧＨ（たとえば、９５％）以上ならば、最終デューティ比ＤＯＵＴＰＧＣに、該上限値ＤＯＵＴＰＧＨをセットする（Ｓ１２２）。ステップＳ１２０およびＳ１２１において、デューティ比ＰＧＣＭＤが、上限値ＤＯＵＴＰＧＨおよび予め決められた下限値ＤＯＵＴＰＧＬ（たとえば、５％）の間の値を持つならば、最終デューティ比ＤＯＵＴＰＧＣに、デューティ比ＰＧＣＭＤをセットする（Ｓ１２３）。
【０１００】
ステップＳ１２１において、デューティＰＧＣＭＤが下限値ＤＯＵＴＰＧＬ以下ならば、最終デューティ比ＤＯＵＴＰＧＣにゼロを設定する（Ｓ１２４）。この場合、最終デューティ比ＤＯＵＴＰＧＣがゼロなので、デューティレートＰＧＲＡＴＥおよびパージ流量レートＱＲＡＴＥにゼロが設定され、高濃度補正係数ＫＫＥＶＧに値１がセットされ、漸増係数ＫＥＶＡＣＴがゼロにセットされる（Ｓ１２５）。ステップＳ１２６に進み、前述した式（１１）に従ってデューティレートＰＧＲＡＴＥを求める。
【０１０１】
図５は、図４のステップＳ１１１で実行される、パージ量ＱＰＧＣを求めるフローチャートである。ステップＳ１５１において、パージ流量係数ＫＰＧＴを求める。パージ流量係数ＫＰＧＴは、運転状態に応じて算出される。
【０１０２】
ステップＳ１５２〜Ｓ１５７は、目標パージ流量ＱＰＧＣＭＤを求める処理である。ステップＳ１５２において、前述した式（６）に従い、基本パージ流量ＱＰＧＣＢＡＳＥに係数ＫＰＧＴを乗算した値を、一時変数ｑｐｇｃｍｄにセットする。一次変数ｑｐｇｃｍｄが、予め設定された上限値ＱＰＧＭＡＸ（たとえば、３０L/min）より大きければ、該上限値がパージ流量ＱＰＧＣＭＤに設定される（Ｓ１５３およびＳ１５７）。一次変数ｑｐｇｃｍｄが、予め設定された下限値（たとえば１L/min）より小さければ、該下限値が目標パージ流量ＱＰＧＣＤＭＤにセットされる（Ｓ１５４およびＳ１５５）。一次変数ｑｐｇｃｍｄが上限値および下限値の間にあるならば、該一時変数ｑｐｇｃｍｄにセットされた値が目標パージ流量ＱＰＧＣＭＤに設定される（Ｓ１５６）。
【０１０３】
ステップＳ１５８〜Ｓ１６７は、前述した式（７）に従ってパージ流量ＱＰＧＣを求める処理である。前述したように、パージ流量ＱＰＧＣは、段階的に目標パージ流量ＱＰＧＣＭＤに達するように制御される。ステップＳ１５８において、パージカットフラグＦ＿ＰＧＲＥＱに１が設定されているかどうかを調べる。１が設定されていれば、パージカットが実行されていること（すなわち、パージされていない状態）を示すので、前回のサイクルで算出されたパージ流量ＱＰＧＣ（ｋ−１）から所定値ＤＱＰＧＣＯＢＤ（たとえば、２L/min）だけ減らした値を、一時変数ｑｐｇｃに代入する（Ｓ１６４）。
【０１０４】
パージカットフラグがゼロならば、一時変数ｄｑｐｇｃに、「吸入空気量ＱＡＩＲ×パージ加算係数ＫＤＱＰＧＣ」を代入する。パージ加算係数ＫＤＱＰＧＣは、吸入空気量ＱＡＩＲに対してどれくらいをパージ流量として加算するかを定める係数（たとえば、０．００３）である。ステップＳ１６０おいて、一時変数ｄｑｐｇｃの値が、予め決められた上限値ＤＱＰＧＣＭＡＸ（たとえば、２L/min）以上ならば、該上限値をデルタパージ量ＤＱＰＧＣにセットし（１６２）、該上限値ＤＱＰＧＣＭＡＸより小さければ、一時変数ｄｑｐｇｃの値をデルタパージ量ＤＱＰＧＣにセットする（Ｓ１６１）。ステップＳ１６３に進み、前回のサイクルで算出されたパージ流量ＱＰＧＣ（ｋ−１）に、ステップＳ１６１またはＳ１６２で求めたデルタパージ量ＤＱＰＧＣを加算する。こうして、目標パージ流量ＱＰＧＣＭＤに向けて、パージ流量が増やされる。
【０１０５】
ステップＳ１６５では、一時変数ｑｐｇｃが目標パージ流量ＱＰＧＣＭＤを超えたかどうか判断する。超えたならば、目標パージ流量ＱＰＧＣＭＤの値をパージ流量ＱＰＧＣにセットし（Ｓ１６６）、超えなければ、一時変数ｑｐｇｃの値をパージ流量ＱＰＧＣにセットする（Ｓ１６７）。こうして、今回のサイクルにおいてパージされるパージ流量ＱＰＧＣが、目標パージ流量ＱＰＧＣＭＤを超えないよう算出される。
【０１０６】
ステップＳ１６８において、漸増係数ＫＥＶＡＣＴを、所定値ＤＫＥＶＡＣＴだけ増やす。ステップＳ１６９および１７０において、漸増係数ＫＥＶＡＣＴは、値１以下にリミット処理される。こうして、漸増係数ＫＥＶＡＣＴは、前述した図４のステップＳ１２１においてデューティ比ＰＧＣＭＤが下限値ＤＯＵＴＰＧＬ以下であってパージカット状態であるときにはゼロに初期化され、ＰＧＣＭＤ＞ＤＯＵＴＰＧＬとなってパージが開始された時点より該ルーチンが実行されるたびに、値１に向けて徐々に増やされる。
【０１０７】
図６は、空燃比補正係数ＫＡＦを算出するフローチャートである。この実施例では、空燃比補正係数ＫＡＦはＰＩＤ制御によって算出される。しかし、他の制御方法によって算出するようにしてもよい。このフローチャートは、たとえばＴＤＣセンサからＴＤＣパルス信号が出力されるたびに実行される。
【０１０８】
ステップＳ２０１において、現在空燃比フィードバック制御が行われているかどうかを判断する。現在空燃比フィードバック制御が行われていなければ、オープンループ制御であることを示す。この場合ステップＳ２０２に進み、空燃比補正係数ＫＡＦに値１．０をセットする。
【０１０９】
空燃比フィードバック制御が行われていれば、前回のサイクルにおいて空燃比フィードバック制御が行われていたかどうかを判断する（Ｓ２０３）。前回のサイクルでフィードバック制御が実行されていなかったならば、ステップＳ２０４およびＳ２０５でＰＩＤ制御のための初期値をセットする。具体的には、今回のサイクルの積分項ＫＩＦ(k)に、最新の空燃比係数ＫＡＦをセットし（Ｓ２０４）、変数ＤＫＣＭＤＢに、目標空燃比係数ＫＣＭＤに対する実空燃比係数ＫＡＣＴの偏差ＤＫＣＭＤをセットする（Ｓ２０５）。
【０１１０】
ステップＳ２０６で、ＰＩＤ制御の比例項ＫＰＦを算出する。比例項ＫＰＦは、実験的に決定される比例制御ゲイン＃ＫＰＡＦに、偏差ＤＫＣＭＤを乗算して算出される。ステップＳ２０７では、積分項ＫＩＦを算出する。積分項ＫＩＦは、実験的に決定される積分制御ゲイン＃ＫＩＡＦに偏差ＤＫＣＭＤを乗算したものを、前回のサイクルで算出された積分項ＫＩＦ(k-1)に加算することにより算出される。ステップＳ２０８では、ＰＩＤ制御の微分項ＫＤＦを算出する。微分項ＫＤＦは、実験的に決定される微分制御ゲイン＃ＫＤＡＦに、今回のサイクルにおける偏差ＤＫＣＭＤと前回のサイクルにおける偏差ＤＫＣＭＤＢとの差を乗算することにより、算出される。
【０１１１】
ステップＳ２０９に進み、比例項ＫＰＦ、積分項ＫＩＦ、および微分項ＫＤＦを加算し、空燃比補正係数ＫＡＦを算出する。ステップＳ２１０に進み、今回のサイクルにおける偏差ＤＫＣＭＤを変数ＤＫＣＭＤＢにシフトする。
【０１１２】
ステップＳ２１１において、現在パージが実行されているかどうか判断する。パージが実行されていなければ、前述した式（９）に従い、パージ無し学習値ＫＲＥＦＸを算出する（Ｓ２１２）。パージが実行されていれば、前述した式（９）に従い、パージ有り学習値ＫＲＥＦを算出する（Ｓ２１３）。
【０１１３】
図７は、パージ補正係数ＫＡＦＥＶＡＣＴを算出するフローチャートである。このフローチャートは、たとえばＴＤＣセンサからＴＤＣパルス信号が出力されるたびに実行される。
【０１１４】
ステップＳ３０１において、目標パージ補正係数ＫＡＦＥＶＡＣＺを求めるルーチン（図８）を実行する。目標パージ補正係数ＫＡＦＥＶＡＣＺは、前述したように、高濃度補正係数ＫＫＥＶＧで補正される前の、要求燃料に対する蒸発燃料の割合を示す係数である。
【０１１５】
ステップＳ３０２〜Ｓ３０４は、高濃度補正係数ＫＫＥＶＧを算出する処理を示す。ステップＳ３０２において、ベーパ濃度係数ＫＡＦＥＶと、ガード係数ＫＥＶＡＣＴＧを比較する。ガード係数ＫＥＶＡＣＴＧは、運転状態に応じて算出される係数である。ベーパ濃度係数ＫＡＦＥＶがガード係数ＫＥＶＡＣＴＧより大きければ、蒸発燃料の濃度が非常に濃いことを示す。この場合、「ＫＡＦＥＶ／ＫＥＶＡＣＴＧ」を計算し、高濃度係数ＫＫＥＶＧを求める（Ｓ３０３）。一方、ベーパ濃度係数ＫＡＦＥＶがガード係数ＫＥＶＡＣＴＧより小さければ、蒸発燃料の濃度が補正するほど大きくないことを示す。この場合、高濃度補正係数ＫＫＥＶＧに１をセットする（Ｓ３０４）。
【０１１６】
ステップＳ３０５に進み、前述の式（１４）に示されるように、目標パージ補正係数ＫＡＦＥＶＡＣＺに、ステップＳ３０３またはＳ３０４で求めた高濃度係数ＫＫＥＶＧと、ステップＳ１６８〜Ｓ１７０（図５）で求めた漸増係数ＫＥＶＡＣＴを乗算し、暫定パージ補正係数ＫＡＦＥＶＡＣを求める。
【０１１７】
ステップＳ３０６において、リーク判定の実施についての許可フラグＦ＿ＥＶＰＬＫＭの値を調べる。フラグＦ＿ＥＶＰＬＫＭが、リーク判定が実施されていないことを示す値ゼロを持つならば、ステップＳ３０７に進み、前述の式（１３）に従ってパージ補正係数ＫＡＦＥＶＡＣＴを求める。フラグＦ＿ＥＶＰＬＫＭが、リーク判定が実施されていることを示す値１を持つならば、ステップＳ３０８に進み、前述の式（１５）に従ってパージ補正係数ＫＡＦＥＶＡＣＴを求める。こうして、リーク判定が実施されないときは、パージ補正量は、算出された蒸発燃料の量に向かって徐々に増えるよう制御され、リーク判定が実施されるときは、パージ補正量は、算出された蒸発燃料の量に最初から設定される。
【０１１８】
ステップＳ３０９に進み、フラグＦ＿ＷＯＴの値を調べる。フラグＦ＿ＷＯＴに値１がセットされていれば、空燃比をリッチ状態にする制御モード（リッチ制御モードと呼ぶ）にあることを示す。フラグＦ＿ＷＯＴは、たとえばスロットルが全開であるとき、触媒が高温であるとき等に値１がセットされる。リッチ制御モードに入ると、オープンループ制御が実行される。
【０１１９】
フラグＦ＿ＷＯＴが値１ならば、パージ無し学習値ＫＲＥＦＸとパージ有り学習値ＫＲＥＦを比較する（Ｓ３１０）。前述したように、パージ無し学習値ＫＲＥＦＸとパージ有り学習値ＫＲＥＦの偏差は、フィードバック制御時に空燃比補正係数ＫＡＦによって修正されていたパージ補正係数ＫＡＦＥＶＡＣＴの“ずれ”に相当する。
【０１２０】
ステップＳ３１０で、パージ有り学習値ＫＲＥＦがパージ無し学習値ＫＲＥＦＸより大きければ、空燃比が目標空燃比に対してリーン状態にあることを示す。したがって、ステップＳ３１１において、ステップＳ３０７またはＳ３０８で求めたパージ補正係数ＫＡＦＥＶＡＣＴを、前述の式（１６）に従って補正する。こうして、空燃比フィードバック制御時に生じていたパージ補正係数ＫＡＦＥＶＡＣＴのずれが解消され、パージ補正係数ＫＡＦＥＶＡＣＴは減る方向に修正される。
【０１２１】
ステップＳ３１０でパージ有り学習値ＫＲＥＦがパージ無し学習値ＫＲＥＦＸ以下ならば、空燃比が目標空燃比に対してリッチ状態にあることを示す。この実施例では、リッチ制御の場合、空燃比がさらにリッチの方向に向かうのは問題なしとして、パージ補正係数ＫＡＦＥＶＡＣＴを補正しない。
【０１２２】
代替的に、ステップＳ３１０でパージ有り学習値ＫＲＥＦがパージ無し学習値ＫＲＥＦＸ以下である場合に、内燃機関に供給される燃料量が要求燃料量に対して過剰とならないようパージ補正係数ＫＡＦＥＶＡＣＴを補正してもよい。この場合、「ＫＡＦＥＶＡＣＴ←ＫＡＦＥＶＡＣＴ＋（ＫＲＥＦＸ−ＫＲＥＦ）×ＰＧＲＡＴＥ×ＱＲＡＴＥ」を実行する。こうすることにより、パージ補正係数ＫＡＦＥＶＡＣＴが増える方向に修正され、要求燃料に適合した量の燃料が内燃機関に供給される。
【０１２３】
図８は、図７のステップＳ３０１で実行される、目標パージ補正係数ＫＡＦＥＶＡＣＺを算出するフローチャートを示す。ステップＳ３５１において、パージ流量レートＱＲＡＴＥを求める。これは、前述した式（１２）に従って算出される。
【０１２４】
ステップＳ３５２において、パージ輸送遅れテーブルＣＰＧＤＬＹＲＸをアクセスし、エンジン回転数ＮＥに基づいてパージ輸送遅れＣＰＧＤＬＹＲＸを求める。前述したように、パージ輸送遅れＣＰＧＤＬＹＲＸは、蒸発燃料がパージ制御弁を介してパージ通路にパージされてから、エンジンの吸気系に到達するまでの時間的遅れを示す。この実施例では、パージ輸送遅れＣＰＧＤＬＹＲＸは整数ｎで表される。図１０に、パージ輸送遅れテーブルＣＰＧＤＬＹＲＸテーブルの例を示す。該テーブルに示されるように、回転ＮＥが大きくなるほど、パージ輸送遅れＣＰＧＤＬＹＲＸも大きくなる。これは、エンジン回転数ＮＥが大きいほど、蒸発燃料がパージされてからエンジンの吸気系に到達するまでの間に実行される、内燃機関における吸入工程数が増加するからである。
【０１２５】
ステップＳ３５３において、リングバッファを１つだけシフトする。リングバッファは、たとえばＫＡＦＥＶＲＴ０〜ＫＡＦＥＶＲＴｆの１６個のバッファから構成される。リングバッファをシフトするということは、ＫＡＦＥＶＲＴ０〜ＫＡＦＥＶＲＴｅに格納されたデータを、ＫＡＦＥＶＲＴ１〜ＫＡＦＥＶＲＴｆにそれぞれシフトすることを示す。ＫＡＦＥＶＲＴｆに格納されていたデータを廃棄される。こうして、ＫＡＦＥＶＲＴ０を空にする。
【０１２６】
ステップＳ３５４において、パージ流量レートＱＲＡＴＥに、図４のステップＳ１２６で算出されたデューティレートＰＧＲＡＴＥを乗算したものを、バッファＫＡＦＥＶＲＴ０に格納する。こうして、ゼロ番目のバッファには、今回のサイクルで算出された「ＰＧＲＡＴＥ×ＱＲＡＴＥ」が格納される。１番目、２番目、．．．のバッファ（すなわち、ＫＡＦＥＶＲＴ１、ＫＡＦＥＶＲＴ２、、、）には、それぞれ、前回、前々回、．．．のサイクルで算出された「ＰＧＲＡＴＥ×ＱＲＡＴＥ」が格納されている。
【０１２７】
ステップＳ３５５に進み、前述の式（１０）に従い、目標パージ補正係数ＫＡＦＥＶＡＣＺを求める。具体的には、ベーパ濃度係数ＫＡＦＥＶに、ステップＳ３５２で求めたパージ輸送遅れの値ｎに対応するＫＡＦＥＶＲＴｎを乗算し、目標パージ補正係数ＫＡＦＥＶＡＣＺを算出する。たとえば、パージ輸送遅れＣＰＧＤＬＹＲＸが「４」ならば、バッファＫＡＦＥＶＲＴ４に格納されたＫＡＦＥＶＲＴの値が使用される。
【０１２８】
図９は、ベーパ濃度係数ＫＡＦＥＶを求めるフローチャートである。このフローチャートは、一定の時間間隔（たとえば１０ミリ秒）ごとに繰り返し実行される。
【０１２９】
ステップＳ４０１において、現在空燃比フィードバック制御が実行されているかどうか判断する。フィードバック制御中でなければこのルーチンを抜け、フィードバック制御中ならばステップＳ４０２に進む。ステップＳ４０２において、パージ流量ＱＰＧＣがゼロならば、今回のサイクルでは蒸発燃料をパージしないことを示すので、このルーチンを抜ける。パージ流量ＱＰＧＣがゼロでなければ、ステップＳ４０３に進む。
【０１３０】
ステップＳ４０３およびＳ４０４において、ＤＫＡＦＥＶＸＨテーブルおよびＤＫＡＦＥＶＸＬテーブルをアクセスし、吸入空気量ＱＡＩＲに基づいて高側判定値ＤＫＡＦＥＶＸＨおよび低側判定値ＤＫＡＦＥＶＸＬをそれぞれ求める。図１１は、このテーブルの例を示す。高側および低側判定値ＤＫＡＦＥＶＸＨおよびＤＫＡＦＥＶＸＬは、吸入空気量ＱＡＩＲが大きくなるほど小さくなるよう設定される。
【０１３１】
ステップＳ４０５において、空燃比学習値ＫＲＥＦＸと低側判定値ＤＫＡＦＥＶＸＬの差が空燃比補正係数ＫＡＦより大きく、かつステップＳ４０６において、実空燃比係数ＫＡＣＴが目標空燃比係数ＫＣＭＤより大きければ、パージによる影響で空燃比が目標空燃比に対してリッチ状態にあることを示す。したがって、ベーパ濃度係数ＫＡＦＥＶを、所定値ＤＫＥＶＡＰＯＰ（たとえば、０．０５）だけ増やす（Ｓ４０７）。
【０１３２】
一方、空燃比学習値ＫＲＥＦＸに高側判定値ＤＫＡＦＥＶＸＨを足したものが、空燃比補正係数ＫＡＦより小さく（Ｓ４０８）、かつ実空燃比係数ＫＡＣＴが目標空燃比係数ＫＣＭＤより小さければ（Ｓ４０９）、空燃比が目標空燃比に対してリーン状態にあることを示す。したがって、ベーパ濃度係数ＫＡＦＥＶを所定値ＤＫＥＶＡＰＯＭ（たとえば、０．０８）だけ減らす（Ｓ４１０）。
【０１３３】
ステップＳ４０５およびＳ４０８の両方の判断ステップがＮｏであるとき、空燃比補正係数ＫＡＦが、高側判定値ＤＫＡＦＥＶＸＨと低側判定値ＤＫＡＦＥＶＸＬの間にあることを示す。この場合、ステップＳ４１１〜Ｓ４１４において、学習値ＫＲＥＦおよびＫＲＥＦＸの差に応じて、ベーパ濃度係数ＫＡＦＥＶを求める。学習値ＫＲＥＦが学習値ＫＲＥＦＸより小さければステップＳ４１２に進み、学習値ＫＲＥＦが学習値ＫＲＥＦＸ以上ならばステップＳ４１４に進む。
【０１３４】
ステップＳ４１２において、空燃比補正係数ＫＡＦが学習値ＫＲＥＦＸより小さければ、パージの影響で空燃比がリッチ側に傾いていることを示す。学習値ＫＲＥＦＸからＫＲＥＦを引いた値に所定値ＣＡＦＥＶ（たとえば、０．０２）を乗算した値を加算することにより、ベーパ濃度係数ＫＡＦＥＶを更新する。この場合、ＫＲＥＦ＜ＫＲＥＦＸなので、ベーパ濃度係数ＫＡＦＥＶは大きくなる。
【０１３５】
一方、ステップＳ４１４において空燃比補正係数ＫＡＦが学習値ＫＲＥＦＸより大きければ、空燃比がリーン側に傾いていることを示す。学習値ＫＲＥＦＸからＫＲＥＦを引いた値に所定値を乗算した値を加算することにより、ベーパ濃度係数ＫＡＦＥＶを更新する。この場合、ＫＲＥＦ＞ＫＲＥＦＸなので、ベーパ濃度係数ＫＡＦＥＶは小さくなる。
【０１３６】
以上、本発明の好ましい実施形態について説明してきた。本発明は、リーク判定を実施するためにパージ制御弁を開いて蒸発燃料排出抑止系を減圧する形態だけでなく、他の目的のために該系を減圧する他の形態にも適用されることができる。
【０１３７】
【発明の効果】
この発明によれば、蒸発燃料排出抑止系についてリーク判定を実施するためにパージ制御弁を大きく開弁して多量の蒸発燃料がパージされた場合でも、燃料噴射量に対する適切な補正を迅速に行うことができ、よって適切な空燃比を実現することができる。
【図面の簡単な説明】
【図１】この発明の一実施例に従う、内燃機関およびその制御装置を概略的に示す図。
【図２】この発明の一実施例に従う、内燃機関の制御装置の機能ブロック図。
【図３】この発明の一実施例に従う、吸入空気量ＱＡＩＲを算出するフローチャート。
【図４】この発明の一実施例に従う、パージ制御弁を駆動するためのデューティＰＧＣＭＤを算出するフローチャート。
【図５】この発明の一実施例に従う、パージ流量ＱＰＧＣを算出するフローチャート。
【図６】この発明の一実施例に従う、空燃比補正係数ＫＡＦを算出するフローチャート。
【図７】この発明の一実施例に従う、パージ補正係数ＫＡＦＥＶＡＣＴを算出するフローチャート。
【図８】この発明の一実施例に従う、目標パージ補正係数ＫＡＦＥＶＡＣＺを算出するフローチャート。
【図９】この発明の一実施例に従う、ベーパ濃度係数ＫＡＦＥＶを算出するフローチャート。
【図１０】この発明の一実施例に従う、エンジン回転数ＮＥに基づくパージ輸送遅れを求めるためのＣＰＧＤＬＹＲＸテーブルの例を示す図。
【図１１】この発明の一実施例に従う、吸入空気量ＱＡＩＲに基づく、空燃比学習値の高側および低側判定値ＤＫＡＦＥＶＸＨおよびＤＫＡＦＥＶＸＬを求めるためのテーブルの例を示す図。
【符号の説明】
１ エンジン ２ 吸気管
５ ＥＣＵ ６ 燃料噴射弁
９ 燃料タンク ２７ パージ通路
３０ パージ制御弁[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a control device for optimizing an air-fuel ratio during a leak determination in an internal combustion engine that determines whether or not there is a leak in a purge system that purges evaporated fuel into an intake system.
[0002]
[Prior art]
Fuel is supplied to an internal combustion engine of an automobile from a fuel tank via a fuel injection valve. On the other hand, the evaporated fuel generated in the fuel tank is adsorbed by the canister, and a part of the evaporated fuel adsorbed by the canister is purged to the intake system of the internal combustion engine through the purge passage. A purge control valve is provided in the purge passage. The amount of evaporated fuel flowing into the internal combustion engine is controlled by adjusting the opening of the purge control valve according to the operating state.
[0003]
Generally, feedback control of air-fuel ratio is performed in an internal combustion engine. In this air-fuel ratio feedback control, the fuel injection amount is calculated so that the air-fuel ratio becomes the target air-fuel ratio, and the fuel injection valve is controlled according to the calculated fuel injection amount. In such air-fuel ratio feedback control, when the evaporated fuel is purged into the intake system of the internal combustion engine, the air-fuel ratio fluctuates. For this reason, there has been proposed a technique for correcting the fuel injection amount so that the amount of evaporated fuel contributing to the fuel supplied to the internal combustion engine is reduced from the fuel injection amount.
[0004]
Japanese Patent Application Laid-Open No. 7-217477 describes a method of adjusting a correction amount for a fuel injection amount. When the purge process is stopped due to a fuel cut or the like, when the fuel supply and the purge process are restarted, the air-fuel ratio may fluctuate. This variation is caused by a delay until the evaporated fuel contributes to the actual fuel injection, the fuel injection amount being corrected by a large correction amount, and the like. In order to avoid the fluctuation of the air-fuel ratio, the correction amount is gradually increased from when the purge process is started.
[0005]
On the other hand, a technique for detecting a leak (leakage) as an abnormality of the purge system from the fuel tank to the intake system is known. In the leak detection method described in Japanese Patent Application Laid-Open No. 9-291854, the purge system is reduced to a predetermined pressure by opening the purge control valve and closing the air release valve (vent shut valve). After setting the purge system to a predetermined negative pressure, the purge system is closed by closing the purge control valve. Whether or not there is a leak in the purge system is determined according to the pressure change for a predetermined period after the purge system is sealed.
[0006]
In such a leak determination, the purge control valve is forcibly set to a relatively large opening in order to quickly bring the purge system into a negative pressure state. Therefore, the amount of evaporated fuel purged to the intake system increases.
[0007]
[Problems to be solved by the invention]
Even if the purge control valve is set to a high opening degree for performing the leak determination, if normal, the correction amount corresponding to the high opening degree is immediately reduced from the fuel injection amount, and the fluctuation of the air-fuel ratio can be suppressed. However, if the leak determination process is started while the correction amount gradual increase control as in the above-described conventional method is being performed, the calculated correction amount increases only gradually, but actually The purge control valve is forcibly opened at a relatively large opening, causing a situation where a large amount of evaporated fuel is purged. Since the fuel injection amount supplied via the fuel injection valve is calculated so as to match the calculated correction amount, as a result, the air-fuel ratio may become rich, and the exhaust gas characteristics and drivability may be impaired.
[0008]
Therefore, there is a need for an apparatus and method that appropriately corrects the fuel injection amount in accordance with the amount of evaporated fuel to be purged when the purge system leak determination is performed, thereby avoiding fluctuations in the air-fuel ratio.
[0009]
[Means for Solving the Problems]
According to one aspect of the present invention, there is provided a control device for an internal combustion engine having a purge system that purges evaporated fuel generated in a fuel tank to an intake system via a purge control valve. The control device includes a fuel amount calculating means (supplied fuel calculating unit 90 and equation (1) in FIG. 2) for calculating a required fuel amount to be supplied to the internal combustion engine based on an operating state of the internal combustion engine, and a purge system. A purge correction amount representing the amount of evaporated fuel purged to the intake system is calculated, and the purge correction amount is corrected so as to reduce the evaporated fuel amount with respect to the required fuel amount calculated by the fuel amount calculation means. Purge correction means (purge correction coefficient calculation unit 89 in FIG. 2, equations (2) and (13)). Further, the control device includes a gradual increase means (purge correction coefficient calculation unit 89 and equation (14) in FIG. 2) for setting the purge correction amount to be gradually increased from a predetermined initial value at the start of the purge, a purge Judgment means (leakage judgment unit 80 in FIG. 2) for judging whether or not an abnormality judgment for judging whether or not the purge system is abnormal is performed based on a pressure change when the control valve is opened and the purge system is set to a negative pressure. ) And prohibiting means (purge correction coefficient calculating unit 89 and equation (15) in FIG. 2) for prohibiting execution of the gradual increase means when it is determined that the abnormality determination is performed by the determining means. .
[0010]
According to the present invention, when the abnormality determination of the purge system is performed, the purge correction amount corresponding to the evaporated fuel is reduced from the required fuel amount without performing the gradual increase control for gradually increasing the purge correction amount. To. Therefore, even when a large amount of evaporated fuel is purged by performing the abnormality determination, an amount of fuel corresponding to the evaporated fuel is injected from the fuel injection valve, so that the air-fuel ratio during the abnormality determination is optimized. Further, deterioration of exhaust gas characteristics and drivability can be avoided.
[0011]
DETAILED DESCRIPTION OF THE INVENTION
Next, an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is an overall configuration diagram of an internal combustion engine and its control device according to an embodiment of the present invention.
[0012]
An electronic control unit (hereinafter referred to as “ECU”) 5 includes a CPU 41 that performs calculations for controlling each part of an internal combustion engine (hereinafter referred to as “engine”) 1, and a program for controlling each part of the engine And a read only memory (ROM) 42 for storing various data, a work area for calculation by the CPU 41, and a random access memory (RAM) for temporarily storing data sent from each part of the engine and control signals sent to each part of the engine 43, an input circuit 44 for receiving data sent from each part of the engine, and an output circuit 45 for sending a control signal to each part of the engine.
[0013]
In FIG. 1, the program is indicated by module 1, module 2, module 3, etc., and the program for controlling the amount of fuel supplied to the internal combustion engine according to the present invention is one of these modules. Included in one or more. Various data used for the calculation are stored in the ROM 42 in the form of table 1, table 2, and the like. The ROM 42 may be a rewritable ROM such as an EEPROM. In this case, the result calculated by the ECU 5 in a certain operation cycle is stored in the ROM and can be used in the next operation cycle.
[0014]
The engine 1 is an engine having, for example, four cylinders, and an intake pipe 2 is connected thereto. A throttle valve 3 is arranged on the upstream side of the intake pipe 2, and a throttle valve opening sensor (θTH) 4 connected to the throttle valve 3 outputs an electrical signal corresponding to the opening of the throttle valve 3. Supply to ECU5.
[0015]
The fuel injection valve 6 is provided for each cylinder in the middle of the intake pipe 2 and between the engine 1 and the throttle valve 3, and the valve opening time is controlled by a control signal from the ECU 5. The fuel supply pipe 7 connects the fuel injection valve 6 and the fuel tank 9, and a fuel pump 8 provided in the middle supplies fuel from the fuel tank to the fuel injection valve 6. A regulator (not shown) is provided between the pump 8 and the fuel injection valve 6 and keeps a differential pressure between the pressure of the air taken in from the intake pipe 2 and the pressure of the fuel supplied through the fuel supply pipe 7 constant. When the fuel pressure is too high, excess fuel is returned to the fuel tank 9 through a return pipe (not shown). Thus, the air taken in through the throttle valve 3 passes through the intake pipe 2 and is mixed with the fuel injected from the fuel injection valve 6 and supplied to the cylinder (not shown) of the engine 1.
[0016]
An intake pipe pressure (PBA) sensor 13 and an intake temperature (TA) sensor 14 are mounted on the downstream side of the throttle valve 3 of the intake pipe 2, and detect the intake pipe pressure PBA and the intake temperature TA, respectively, to generate electric signals. Convert it and send it to the ECU 5. The engine water temperature (TW) sensor 15 is attached to a cylinder peripheral wall (not shown) filled with cooling water in the cylinder block of the engine 1, detects the temperature TW of engine cooling water, and sends this to the ECU 5.
[0017]
A cylinder discrimination (CYL) sensor 34 is mounted around the camshaft or crankshaft (both not shown) of the engine 1 and outputs a cylinder discrimination signal CYL when a specific cylinder reaches a predetermined crank angle position. It is sent to the ECU 5. Similarly, a TDC sensor 36 is mounted around the camshaft or crankshaft and outputs a TDC signal at a predetermined crank angle position related to the piston TDC position. Further, a crank angle (CRK) sensor 38 is attached, and outputs a CRK signal pulse at a cycle of a crank angle (for example, 30 degrees) shorter than the cycle of the TDC signal pulse. The CRK signal pulse is counted by the ECU 5, thereby detecting the engine speed NE.
[0018]
The engine 1 has an exhaust pipe 12 and exhausts through a three-way catalyst 33 which is an exhaust gas purification device provided in the middle of the exhaust pipe 12. The LAF sensor 32 mounted in the middle of the exhaust pipe 12 is a wide area air-fuel ratio sensor, detects the oxygen concentration in the exhaust gas, that is, the actual air-fuel ratio in a range from lean to rich, and sends it to the ECU 5.
[0019]
A spark plug 58 is disposed in a combustion chamber (not shown) of the engine 1, and the spark plug 58 is electrically connected to the ECU 5 via an ignition coil and an igniter 50. Further, a knock sensor 52 is disposed in a cylinder head (not shown) of the engine 1, and outputs a signal according to the vibration of the engine 1 and sends it to the ECU 5.
[0020]
A vehicle speed (VPS) sensor 17 that detects the speed of the vehicle is connected to the ECU 5 and sends the detected vehicle speed signal to the ECU 5. Further, a battery voltage (VB) sensor 18 and an atmospheric pressure (PA) sensor 19 are connected to the ECU 5 to detect the battery voltage and the atmospheric pressure, respectively, and send them to the ECU 5.
[0021]
Input signals from various sensors are passed to the input circuit 44. The input circuit 44 shapes the input signal waveform, corrects the voltage level to a predetermined level, and converts the analog signal value into a digital signal value. The CPU 41 processes the converted digital signal, executes a calculation according to a program stored in the ROM 42, and generates a control signal to be sent to the actuator of each part of the vehicle. This control signal is sent to the output circuit 45, which sends the control signal to the fuel injection valve 6, the igniter 50 and other actuators.
[0022]
Next, an evaporative fuel discharge suppression system (hereinafter simply referred to as a purge system) 31 will be described. The purge system 31 includes a fuel tank 9, a charge passage 20, a canister 25, a purge passage 27, and some control valves, and controls the discharge of evaporated fuel from the fuel tank 9.
[0023]
The fuel tank 9 is connected to the canister 25 via the charge passage 20 so that the evaporated fuel from the fuel tank 9 can move to the canister 25. The charge passage 20 has a first branch 20a and a second branch 20b, which are provided in the engine room. The internal pressure sensor 11 is attached to the fuel tank side of the charge passage 20 and detects a differential pressure between the internal pressure in the charge passage 20 and the atmospheric pressure.
[0024]
A two-way valve 23 is provided in the first branch 20a, and the two-way valve 23 includes two mechanical valves 23a and 23b. The valve 23a is a positive pressure valve that opens when the tank internal pressure becomes a predetermined amount higher than the atmospheric pressure. When this pressure is open, the evaporated fuel flows into the canister 25 and is adsorbed there. The valve 23 b is a negative pressure valve that opens when the tank internal pressure becomes a predetermined amount lower than the pressure on the canister 25 side. When this valve is open, the evaporated fuel adsorbed by the canister 25 returns to the fuel tank 9. The second branch 20b is provided with a bypass valve 24, which is an electromagnetic valve. The bypass valve 24 is normally in a closed state, and opens according to a control signal from the ECU 5.
[0025]
The canister 25 incorporates activated carbon that adsorbs fuel vapor, and has an inlet (not shown) that communicates with the atmosphere via a passage 26a. A vent shut valve 26, which is an electromagnetic valve, is provided in the middle of the passage 26a. The vent shut valve 26 is normally in an open state, and opens according to a control signal from the ECU 5.
[0026]
The canister 25 is connected to the downstream side of the throttle valve 3 in the intake pipe 2 via the purge passage 27. A purge control valve 30, which is an electromagnetic valve, is provided in the middle of the purge passage 27, and the fuel adsorbed by the canister 25 is appropriately purged to the engine intake system via the purge control valve 30. The purge control valve 30 continuously controls the purge flow rate by changing the on-off duty ratio based on a control signal from the ECU 5.
[0027]
Hereinafter, a portion on the fuel tank side from the bypass valve 24 is referred to as a tank system, and a portion on the canister side from the bypass valve 24 is referred to as a canister system.
[0028]
In order to help the understanding of the invention, first, an outline of the required fuel and the purge correction amount will be described. The required fuel is a fuel to be supplied to each cylinder of the engine. The required fuel is calculated by multiplying the basic fuel amount TIM by a fuel injection coefficient (KTOTAL × KCMD × KAF) as shown in the equation (1).
[0029]
[Expression 1]
Required fuel = TIM x (KTOTAL x KCMD x KAF) (1)
[0030]
Here, the basic fuel amount TIM is specifically represented by a basic fuel injection time determined according to the engine speed NE and the intake pipe pressure PBA. KTOTAL is a correction coefficient calculated based on detection signals from various sensors, and is set so that the fuel consumption characteristics and acceleration characteristics of the engine are optimized in accordance with the driving state. KCMD is called a target air-fuel ratio coefficient, and represents the target air-fuel ratio as an equivalent ratio. The target air-fuel ratio coefficient KCMD is proportional to the reciprocal of the air-fuel ratio A / F, that is, the fuel-air ratio F / A, and takes a value of 1.0 when the stoichiometric air-fuel ratio. KAF represents an air-fuel ratio correction coefficient. The air-fuel ratio correction coefficient KAF is a coefficient used for feedback control of the air-fuel ratio so that the air-fuel ratio of the air-fuel mixture supplied to the engine matches the target air-fuel ratio. The air-fuel ratio correction coefficient KAF is calculated based on the actual air-fuel ratio detected by the LAF sensor 32.
[0031]
The amount of fuel to which the purge flow rate QPGC contributes to the required fuel (hereinafter referred to as purge correction amount) is calculated by multiplying the basic fuel amount TIM by the purge correction coefficient KAFEVACT, as shown in equation (2). The Here, the purge flow rate QPGC indicates the flow rate of the evaporated fuel purged into the engine intake system.
[0032]
[Expression 2]
Purge correction amount = TIM × KAFEVACT (2)
[0033]
The purge correction coefficient KAFEVACT is expressed as a ratio to the fuel injection coefficient. The purge correction coefficient KAFEVACT is calculated based on the purge flow rate QPGC and the evaporated fuel concentration (hereinafter referred to as vapor concentration) KAFEV. A detailed calculation method of the purge correction coefficient KAFEVACT will be described later.
[0034]
From the expressions (1) and (2), the fuel injection amount TCYL actually injected through the fuel injection valve is derived as the following expression (3). In this way, an amount of fuel reduced from the required fuel amount by the purge correction amount based on the purge correction coefficient KAFEVACT is supplied from the fuel injection valve.
[0035]
[Equation 3]

[0036]
FIG. 2 is a functional block diagram of the control device according to the present invention. The function represented by each functional block is typically executed by a computer program. Alternatively, each functional block may be realized by arbitrary hardware configured to execute the function represented by each functional block.
[0037]
The leak determination unit 80 determines whether there is a leak in the purge system 31 when a predetermined condition is satisfied. The predetermined conditions are: 1) the purge process is being executed, 2) the engine operating state is in a predetermined steady state, 3) the change in vehicle speed is small, and 4) the air-fuel ratio correction coefficient is in the predetermined value range. , Etc. Alternatively, other conditions may be used, or different conditions may be added. The leak determination unit 80 sets 1 to the flag F_EVPLKM in order to permit execution of the leak determination when a predetermined condition is satisfied.
[0038]
When the flag F_EVPLKM is set to 1, the leak determination unit 80 performs a leak determination of the tank system and the canister system of the purge system 31 according to the following procedure.
[0039]
[Tank system leak judgment]
1) The bypass valve 24 and the vent shut valve 26 are opened, the purge control valve 30 is closed, and the fuel tank 9 is opened to the atmosphere.
[0040]
2) Close the bypass valve 24 and the purge control valve 30, and implement a correction check mode in which the amount of pressure fluctuation per unit time rising from atmospheric pressure to positive pressure is measured as a correction value.
[0041]
3) Open the bypass valve 24, close the vent shut valve 26, and control the purge control valve 30, and implement a pressure reduction mode in which the tank system is stably decompressed to a predetermined pressure.
[0042]
4) After the tank system reaches a predetermined negative pressure state, all the valves 24, 26 and 30 are closed, and a leak check mode is performed in which the amount of pressure fluctuation per unit time to return is measured.
[0043]
5) A value obtained by multiplying the pressure fluctuation amount per unit time in the correction check mode by a predetermined coefficient is subtracted from the pressure fluctuation amount per unit time in the leak check mode. Based on the calculated value, the leak of the tank system is determined. If the calculated value is small, the negative pressure state of the tank system is almost maintained, and it is determined that there is no leak in the tank system. If the calculated value is large, it is determined that there is a leak in the tank system.
[0044]
[Canister system]
1) The bypass valve 24 and the vent shut valve 26 are opened, the purge control valve 30 is closed, and the canister system is opened to the atmosphere.
[0045]
2) The vent shut valve 26 is closed, the purge control valve 30 is opened, and the canister system is stably depressurized to a predetermined pressure.
[0046]
3) After the canister system reaches a predetermined negative pressure state, all the valves 24, 26 and 30 are closed and the pressure fluctuation amount to be returned is measured. If the fluctuation amount is small, it is determined that the negative pressure state of the canister system is almost maintained and there is no leak. If the fluctuation amount is large, it is determined that there is a leak in the canister system.
[0047]
Of course, the present invention can also be applied to the case where the leak determination of either the tank system or the canister system is performed. In addition, as a technique for performing leak determination by setting the target system to a negative pressure, a technique other than the above may be used.
[0048]
The functions shown in the function blocks 81 to 89 realize a technique for correcting the fuel injection amount in accordance with the amount of evaporated fuel to be purged.
[0049]
The intake air amount calculation unit 81 calculates the intake air amount QAIR according to the equation (4).
[0050]
[Expression 4]
QAIR = TIM × NE × 2 × KQAIR × KPA (4)
[0051]
Here, TIM indicates the basic fuel injection amount described above. KQAIR is a coefficient for converting the fuel injection amount into the air flow rate, and has a fixed value (for example, 0.45 L / ms). KPA is a coefficient for correcting the fluctuation of the flow rate according to the intake pipe pressure PBA.
[0052]
The intake air amount calculation unit 81 further calculates the ratio of the evaporated fuel to the intake air amount QAIR, that is, the basic purge amount QPGCBASE according to the equation (5).
[0053]
[Equation 5]
QPGCBASE = QAIR × KQPGB (5)
[0054]
Here, KQPGB indicates a target purge rate, for example, 0.04. In this case, the evaporated fuel is included in 4% of the intake air amount QAIR.
[0055]
The purge flow rate calculation unit 82 calculates the target purge flow rate QPGCMD according to the equation (6) based on the basic purge amount QPGCBASE. The target purge flow rate QPGCMD represents the target value of the purge flow rate purged into the engine intake system in the current cycle.
[0056]
[Formula 6]
QPGCMD = QPGCBASE × KPGT (6)
[0057]
Here, KPGT is a purge flow coefficient and has a value of 1 or less. The amount of the target purge flow rate QPGCMD can be controlled using the coefficient KPGT. The coefficient KPGT is calculated according to the operating state.
[0058]
Further, the purge flow rate calculation unit 82 calculates the purge flow rate QPGC purged in the current cycle based on the intake air amount QAIR according to the equation (7).
[0059]
[Expression 7]
QPGC (k) = QPGC (k−1) + (QAIR × KDQPGC) (7)
[0060]
Here, k is a number for identifying the cycle, (k) indicates the current cycle, and (k-1) indicates the previous cycle. KDQPGC is a predetermined fixed value (for example, 0.003). As apparent from the equation (7), the purge flow rate QPGC is controlled so as to gradually reach the target purge flow rate QPGCMD.
[0061]
The duty calculator 83 calculates a duty ratio PGCMD for driving the purge control valve according to the equation (8) so that the purge flow QPGC received from the purge flow calculator 82 is purged. The duty ratio represents a ratio at which the purge control valve is opened.
[0062]
[Equation 8]

[0063]
KDUTY is a coefficient for converting the purge flow rate into a duty ratio, and has a fixed value (for example, 3.8% · min / L). Since the opening degree of the purge control valve changes according to the differential pressure, KDPBG is a coefficient for correcting it. PGCMD0 represents a duty ratio corresponding to the purge flow rate QPGC, and is hereinafter referred to as a target duty ratio. Since some delay occurs until the purge control valve starts to open depending on the battery voltage VB and the intake pipe pressure PBA, DPGCVBX and DPGC0 are coefficients for correcting this delay (hereinafter referred to as invalid time).
[0064]
The duty calculation unit 83 performs limit processing on the duty ratio PGCMD with a predetermined upper limit value and lower limit value, and outputs a final duty ratio DOUTPGC. Thus, the opening degree of the purge control valve is controlled according to the final duty ratio DOUTPGC.
[0065]
The transport delay calculation unit 84 calculates the purge transport delay CPGDLYRX based on the engine speed NE. The purge transport delay CPGDLYRX indicates a time delay from when the evaporated fuel is purged to the purge passage until it is sent to the intake system of the engine. The purge transport delay CPGDLYRX is represented by an integer n, and indicates that the transport delay increases as n increases. Alternatively, the purge transport delay may be calculated based on the intake air amount QAIR instead of the engine speed NE.
[0066]
Based on the output from the LAF sensor 32, the air-fuel ratio controller 85 calculates an air-fuel ratio coefficient KAF for causing the air-fuel ratio of the air-fuel mixture supplied to the internal combustion engine to converge to the target air-fuel ratio by feedback control.
[0067]
The air-fuel ratio learning value calculation unit 86 calculates the purged learning value KREF and the purgeless learning value KREFX based on the following equation (9) based on the air-fuel ratio coefficient KAF calculated by the air-fuel ratio control unit 85. calculate. As described above, (k) indicates the current cycle, and (k-1) indicates the previous cycle.
[0068]
[Equation 9]

[0069]
Here, #CREF and #CREFX are predetermined values each representing a weighting coefficient. This value determines how much the air-fuel ratio correction coefficient KAF is reflected to KREF (k-1) and KREFX (k-1) calculated in the previous cycle.
[0070]
The vapor concentration coefficient calculation unit 87 calculates the vapor concentration coefficient KAFEV based on the air-fuel ratio correction coefficient KAF calculated by the air-fuel ratio control unit 85 and the learning values KREFX and KREF calculated by the air-fuel ratio learning value calculation unit 86. .
[0071]
The target purge correction coefficient calculation unit 88 calculates the target purge correction coefficient KAFEVACZ according to the following equation (10).
[0072]
[Expression 10]

[0073]
PGRATE and QRATE are expressed by equations (11) and (12).
[0074]
[Expression 11]

[Expression 12]
QRATE = QPGC / QPGCBASE (12)
[0075]
“PGRATE × QRATE” indicates the ratio of the purge flow rate purged to the engine intake system in the current cycle. More specifically, since DOUTPGC represents the final duty ratio as described above, the duty rate PGRATE shown in Expression (11) is an actual duty ratio (invalid time) with respect to the target duty ratio PGCMD0 of the purge flow rate QPGC. The ratio is calculated by subtracting. Therefore, “PGRATE × QPGC” indicates the actual purge flow rate purged by the purge control valve in the current cycle.
[0076]
On the other hand, the basic purge flow rate QPGCBASE indicates the flow rate of the evaporated fuel that can be included in the intake air amount QAIR in the current cycle, as described above with reference to the equation (5). Since the fuel is expressed depending on the vapor concentration coefficient KAFEV, the target purge correction coefficient KAFEVACZ can be obtained by multiplying “PGRATE × QRATE” by the vapor concentration coefficient KAFEV as shown in the equation (10). it can.
[0077]
The target purge correction coefficient calculation unit 88 has a plurality of buffers assigned numbers 0 to (n−1), and stores the target purge correction coefficient calculated in the current cycle in the zeroth buffer. The target purge correction coefficient calculated in the previous cycle is stored in the first buffer,. . . Thus, the target purge correction coefficient is stored in time series. The target purge correction coefficient calculation unit 88 receives the value n of the transport delay CPGDLYRX of the evaporated fuel from the transport delay calculation unit 84, and extracts the target purge correction coefficient KAFEVACZ from the numbered buffer corresponding to the n. For example, if the value n of CPGDLYRX received is 3, the target purge correction coefficient KAFEVACZ is extracted from the third buffer.
[0078]
Based on the calculated target purge correction coefficient KAFEVACZ, the purge correction coefficient calculation unit 89 refers to a purge correction coefficient KAFEVACT (expression (2)) that indicates the ratio of the fuel amount that the purge flow rate QPGC contributes to the required fuel. As described above).
[0079]
The purge correction coefficient calculation unit 89 calculates the purge correction coefficient KAFEVACT according to the equations (13) and (14) when the leak determination is not performed by the leak determination unit 80 (that is, when the flag F_EVPLKM is zero).
[0080]
[Formula 13]

[Expression 14]

[0081]
Here, KKEVG indicates a high concentration correction coefficient. If the concentration of the evaporated fuel is very high, the influence on the air-fuel ratio is large. Therefore, KKEVG is a coefficient for correcting this. The high concentration correction coefficient KKEVG is set to a value larger than 1 accordingly when the concentration of the evaporated fuel is very high.
[0082]
The gradually increasing coefficient KEVACT is a coefficient that is gradually increased from a predetermined initial value (for example, zero) to 1.0. The gradual increase coefficient KEVACT is set to an initial value when the purge is started, and thereafter is increased by a predetermined value in each cycle. Thus, by gradually increasing the purge correction amount that contributes to the required fuel amount, fluctuations in the air-fuel ratio when starting the purge process are prevented. During the purge cut, the gradual increase coefficient KEVACT is set to zero.
[0083]
#CKAFEVAC is a predetermined value representing a weighting coefficient. This value determines how much the provisional purge correction coefficient KAFEVAC is reflected to the purge correction coefficient KAFEVACT (k−1) calculated in the previous cycle.
[0084]
The purge correction coefficient calculation unit 89 calculates the purge correction coefficient KAFEVACT according to the equation (15) when the leak determination is performed by the leak determination unit 80 (that is, when the flag F_EVPLKM is 1).
[0085]
[Expression 15]
KAFEVACT = KAFEVACZ × KKEVG Formula (15)
[0086]
As apparent from the comparison with Expression (14), when the leak determination is performed, the gradual increase coefficient KEVACT is not used for calculating the purge correction coefficient KAFEVACT.
[0087]
At the start of purge, fluctuation of the air-fuel ratio can be suppressed by narrowing the ratio of the purge correction amount to the required fuel with the gradual increase coefficient KEVACT. However, as described above, during the leak determination, the purge control valve is forcibly opened to a predetermined opening in order to quickly bring the purge system to a negative pressure. As a result, more evaporated fuel than the calculated purge correction amount is purged, the air-fuel ratio becomes rich, and the operating state may be deteriorated. In order to avoid this, during the leak determination, the gradual increase control of the purge correction amount by the gradual increase coefficient KEVACT is not performed. Thus, during the leak determination, the calculated amount of evaporated fuel is set from the beginning as the purge correction amount, so that the air-fuel ratio is properly maintained.
[0088]
Further, in the open loop control, the purge correction coefficient calculation unit 89 corrects the correction coefficient KAFEVACT calculated based on the above formula (13) or (15) according to the following formula (16).
[0089]
[Expression 16]

[0090]
Here, the purged learning value KREF is a value obtained by averaging the air-fuel ratio correction coefficient KAF while purging is being performed. The purgeless learning value KREFX is a value obtained by averaging the air-fuel ratio coefficient KAF during the purge cut. As shown in Expression (16), the correction coefficient KAFEVACT is corrected by a deviation (KREF−KREFX) between the purged learning value and the purgeless learning value. The learning values KREF and KREFX are those calculated at the time of the most recently executed feedback control.
[0091]
Since the vapor concentration coefficient KAFEV is calculated by estimation, an error may be included depending on the operating state. As a result, a “deviation” occurs in the purge correction coefficient KAFEVACT. While the air-fuel ratio feedback control is being executed, the air-fuel ratio is controlled so as to reach the target air-fuel ratio by the air-fuel ratio correction coefficient KAF, so that the deviation can be corrected.
[0092]
However, while the open loop control is being executed, the air-fuel ratio correction coefficient KAF is fixed at, for example, a value of 1.0, so that the deviation cannot be corrected with the air-fuel ratio correction coefficient KAF. Therefore, during open loop control, the deviation corrected by the air-fuel ratio correction coefficient KAF during feedback control is obtained, and the purge correction coefficient KAFEVACT is corrected by an amount corresponding to the deviation. The difference between the purged learned value KREF and the purgeless learned value KREFX represents this “deviation”. Thus, the deviation of the purge correction coefficient KAFEVACT included in the feedback control is eliminated, and an amount of fuel suitable for the required fuel is supplied to the internal combustion engine even in the open loop control.
[0093]
The supplied fuel calculation unit 90 receives the purge correction coefficient KAFEVACT from the purge correction coefficient calculation unit 89, and calculates the fuel injection amount TCYL according to the above equation (3). The calculated fuel injection amount TCYL is supplied to the engine via the fuel injection valve. Thus, an amount of fuel obtained by subtracting the amount of evaporated fuel (that is, the purge correction amount) contributed by the purge flow rate QPGC is supplied to the engine via the fuel injection valve.
[0094]
FIG. 3 is a flowchart for obtaining the intake air amount QAIR and the basic purge flow rate QPGBASE. This flow is executed each time a TDC pulse signal is output from a TDC sensor, for example. In step S101, the intake air amount QAIR is obtained according to the above-described equation (4). In step S102, the basic purge flow rate QPGCBASE is obtained according to the above-described equation (5).
[0095]
FIG. 4 is a flowchart for obtaining the final duty ratio DOUTPGC for driving the purge control valve. This flow is repeatedly executed at regular time intervals (for example, 80 milliseconds).
[0096]
In step S111, a routine for obtaining the purge flow rate QPGC (FIG. 5) is executed. In step S112, if the purge flow rate QPGC calculated in step S111 is zero, it indicates that the evaporated fuel is not purged, so the duty ratio PGCMD is set to zero (S113), and the process proceeds to step S120. If the purge flow rate QPGC is not zero in step S112, the process proceeds to step S114, and a period (for example, 80 milliseconds) for driving the purge control valve is set.
[0097]
Steps S115 to S119 are processes for obtaining the duty ratio PGCMD according to the above-described equation (8). In step S115, in order to correct the invalid time of the purge control valve according to the battery voltage, the DPGCVBX table is accessed to obtain the invalid time DPGCVBX based on the battery voltage VB. The DPGCVBX table is set so that the invalid time DPGCVBX decreases as the battery voltage increases.
[0098]
Proceeding to step S116, the KDPBG table is accessed to correct the fluctuation in the duty of the purge control valve due to the differential pressure, and the differential pressure correction value KDPBG is obtained based on the intake pipe pressure PBA. The KDPBG table is set so that the differential pressure correction value KDPBG increases as the engine load decreases. In step S117, the DPGC0 table is accessed to correct the invalid time of the purge control valve according to the intake pipe pressure PBA, and the invalid time DPGC0 is obtained based on the intake pipe pressure PBA. The DPGC0 table is set so that the invalid time DPGC0 increases as the load increases. In step S118 and step S119, the duty ratio PGCMD is obtained according to the above equation (8).
[0099]
Steps S120 to S125 show limit processing for the duty ratio PGCMD. In step S120, if the duty ratio PGCMD is greater than or equal to a predetermined upper limit value DOUTPGH (for example, 95%), the upper limit value DOUTPGH is set to the final duty ratio DOUTPGC (S122). If the duty ratio PGCMD has a value between the upper limit value DOUTPGH and a predetermined lower limit value DOUTPGL (for example, 5%) in steps S120 and S121, the duty ratio PGCMD is set to the final duty ratio DOUTPGC ( S123).
[0100]
If the duty PGCMD is equal to or lower than the lower limit value DOUTPGL in step S121, the final duty ratio DOUTPGC is set to zero (S124). In this case, since the final duty ratio DOUTPGC is zero, the duty rate PGRATE and the purge flow rate QRATE are set to zero, the value 1 is set to the high concentration correction coefficient KKEVG, and the gradual increase coefficient KEVACT is set to zero (S125). Proceeding to step S126, the duty rate PGRATE is determined according to the above-described equation (11).
[0101]
FIG. 5 is a flowchart for obtaining the purge amount QPGC, which is executed in step S111 of FIG. In step S151, a purge flow coefficient KPGT is obtained. The purge flow coefficient KPGT is calculated according to the operating state.
[0102]
Steps S152 to S157 are processes for obtaining the target purge flow rate QPGCMD. In step S152, the value obtained by multiplying the basic purge flow rate QPGCBASE by the coefficient KPGT is set in the temporary variable qpgcmd according to the above-described equation (6). If primary variable qpgcmd is larger than a preset upper limit value QPGMAX (for example, 30 L / min), the upper limit value is set to purge flow rate QPGCMD (S153 and S157). If primary variable qpgcmd is smaller than a preset lower limit value (for example, 1 L / min), the lower limit value is set to target purge flow rate QPGCDMD (S154 and S155). If the primary variable qpgcmd is between the upper limit value and the lower limit value, the value set in the temporary variable qpgcmd is set as the target purge flow rate QPGCMD (S156).
[0103]
Steps S158 to S167 are processes for obtaining the purge flow rate QPGC according to the above-described equation (7). As described above, the purge flow rate QPGC is controlled so as to reach the target purge flow rate QPGCMD step by step. In step S158, it is checked whether or not 1 is set in the purge cut flag F_PGREQ. If 1 is set, it indicates that purge cut is being performed (that is, not purged), and therefore, a predetermined value DQPGCOBD (for example, from the purge flow rate QPGC (k−1) calculated in the previous cycle) 2L / min) is substituted for the temporary variable qpgc (S164).
[0104]
If the purge cut flag is zero, “intake air amount QAIR × purge addition coefficient KDQPGC” is substituted for temporary variable dqpgc. The purge addition coefficient KDQPGC is a coefficient (for example, 0.003) that determines how much is added as the purge flow rate to the intake air amount QAIR. In step S160, if the value of temporary variable dqpgc is equal to or greater than a predetermined upper limit value DQPGCMAX (for example, 2 L / min), the upper limit value is set to delta purge amount DQPGC (162) and should be smaller than upper limit value DQPGCMAX. For example, the value of the temporary variable dqpgc is set to the delta purge amount DQPGC (S161). Proceeding to step S163, the delta purge amount DQPGC calculated at step S161 or S162 is added to the purge flow rate QPGC (k-1) calculated in the previous cycle. Thus, the purge flow rate is increased toward the target purge flow rate QPGCMD.
[0105]
In step S165, it is determined whether or not the temporary variable qpgc exceeds the target purge flow rate QPGCMD. If exceeded, the target purge flow rate QPGCMD is set to the purge flow rate QPGC (S166), and if not exceeded, the value of the temporary variable qpgc is set to the purge flow rate QPGC (S167). Thus, the purge flow rate QPGC purged in the current cycle is calculated so as not to exceed the target purge flow rate QPGCMD.
[0106]
In step S168, the gradual increase coefficient KEVACT is increased by a predetermined value DKEVACT. In steps S169 and 170, the gradual increase coefficient KEVACT is limited to a value of 1 or less. Thus, the gradual increase coefficient KEVACT is initialized to zero when the duty ratio PGCMD is equal to or lower than the lower limit value DOUTPGL and is in the purge cut state in step S121 of FIG. Each time the routine is executed, the value is gradually increased toward the value 1.
[0107]
FIG. 6 is a flowchart for calculating the air-fuel ratio correction coefficient KAF. In this embodiment, the air-fuel ratio correction coefficient KAF is calculated by PID control. However, it may be calculated by another control method. This flowchart is executed each time a TDC pulse signal is output from a TDC sensor, for example.
[0108]
In step S201, it is determined whether air-fuel ratio feedback control is currently being performed. If the air-fuel ratio feedback control is not currently being performed, this indicates open-loop control. In this case, the process proceeds to step S202, and the value 1.0 is set to the air-fuel ratio correction coefficient KAF.
[0109]
If air-fuel ratio feedback control has been performed, it is determined whether air-fuel ratio feedback control has been performed in the previous cycle (S203). If feedback control has not been executed in the previous cycle, initial values for PID control are set in steps S204 and S205. Specifically, the latest air-fuel ratio coefficient KAF is set in the integral term KIF (k) of this cycle (S204), and the deviation DKCMD of the actual air-fuel ratio coefficient KACT with respect to the target air-fuel ratio coefficient KCMD is set in the variable DKCMDB. (S205).
[0110]
In step S206, a proportional term KPF for PID control is calculated. The proportional term KPF is calculated by multiplying the proportional control gain #KPAF determined experimentally by the deviation DKCMD. In step S207, an integral term KIF is calculated. The integral term KIF is calculated by adding an experimentally determined integral control gain #KIAF multiplied by the deviation DKCMD to the integral term KIF (k−1) calculated in the previous cycle. In step S208, a differential term KDF for PID control is calculated. The differential term KDF is calculated by multiplying the differential control gain #KDAF determined experimentally by the difference between the deviation DKCMD in the current cycle and the deviation DKCMDB in the previous cycle.
[0111]
In step S209, the proportional term KPF, the integral term KIF, and the derivative term KDF are added to calculate an air-fuel ratio correction coefficient KAF. Proceeding to step S210, the deviation DKCMD in the current cycle is shifted to the variable DKCMDB.
[0112]
In step S211, it is determined whether a purge is currently being executed. If the purge is not executed, the purgeless learning value KREFX is calculated according to the above-described equation (9) (S212). If purging has been executed, the purged learning value KREF is calculated according to the above-described equation (9) (S213).
[0113]
FIG. 7 is a flowchart for calculating the purge correction coefficient KAFEVACT. This flowchart is executed each time a TDC pulse signal is output from a TDC sensor, for example.
[0114]
In step S301, a routine (FIG. 8) for obtaining the target purge correction coefficient KAFEVACZ is executed. As described above, the target purge correction coefficient KAFEVACZ is a coefficient indicating the ratio of the evaporated fuel to the required fuel before being corrected with the high concentration correction coefficient KKEVG.
[0115]
Steps S302 to S304 show processing for calculating the high density correction coefficient KKEVG. In step S302, the vapor concentration coefficient KAFEV is compared with the guard coefficient KEVACTG. The guard coefficient KEVACTG is a coefficient calculated according to the driving state. If the vapor concentration coefficient KAFEV is larger than the guard coefficient KEVACTG, it indicates that the concentration of the evaporated fuel is very high. In this case, “KAFEV / KEVACACTG” is calculated to obtain a high concentration coefficient KKEVG (S303). On the other hand, if the vapor concentration coefficient KAFEV is smaller than the guard coefficient KEVACTG, it indicates that the concentration of the evaporated fuel is not so large as to be corrected. In this case, 1 is set to the high density correction coefficient KKEVG (S304).
[0116]
Proceeding to step S305, as shown in the aforementioned equation (14), the target purge correction coefficient KAFEVACZ is added to the high concentration coefficient KKEVG obtained in step S303 or S304, and the incremental coefficient obtained in steps S168 to S170 (FIG. 5). Multiply by KEVACT to obtain a provisional purge correction coefficient KAFEVAC.
[0117]
In step S306, the value of the permission flag F_EVPLKM regarding the execution of the leak determination is checked. If the flag F_EVPLKM has a value of zero indicating that the leak determination has not been performed, the process proceeds to step S307, and the purge correction coefficient KAFEVACT is obtained according to the above-described equation (13). If the flag F_EVPLKM has a value 1 indicating that the leak determination has been performed, the process proceeds to step S308, and the purge correction coefficient KAFEVACT is obtained according to the above-described equation (15). Thus, when the leak determination is not performed, the purge correction amount is controlled to gradually increase toward the calculated amount of evaporated fuel, and when the leak determination is performed, the purge correction amount is calculated as the calculated evaporation amount. The amount of fuel is set from the beginning.
[0118]
In step S309, the value of the flag F_WOT is checked. If the value 1 is set in the flag F_WOT, it indicates that the control mode for bringing the air-fuel ratio to a rich state (referred to as rich control mode) is in effect. The flag F_WOT is set to a value of 1, for example, when the throttle is fully open or when the catalyst is hot. When the rich control mode is entered, open loop control is executed.
[0119]
If the flag F_WOT is 1, the purgeless learning value KREFX is compared with the purged learning value KREF (S310). As described above, the deviation between the purgeless learning value KREFX and the purged learning value KREF corresponds to the “deviation” of the purge correction coefficient KAFEVACT that has been corrected by the air-fuel ratio correction coefficient KAF during feedback control.
[0120]
In step S310, if the purged learned value KREF is larger than the purgeless learned value KREFX, it indicates that the air-fuel ratio is lean with respect to the target air-fuel ratio. Accordingly, in step S311, the purge correction coefficient KAFEVACT obtained in step S307 or S308 is corrected according to the above-described equation (16). Thus, the deviation of the purge correction coefficient KAFEVACT that has occurred during the air-fuel ratio feedback control is eliminated, and the purge correction coefficient KAFEVACT is corrected to decrease.
[0121]
If the purged learning value KREF is equal to or less than the purgeless learning value KREFX in step S310, it indicates that the air-fuel ratio is in a rich state with respect to the target air-fuel ratio. In this embodiment, in the case of rich control, the purge correction coefficient KAFEVACT is not corrected, assuming that there is no problem that the air-fuel ratio goes further in the rich direction.
[0122]
Alternatively, when the purged learned value KREF is equal to or less than the purged learned value KREFX in step S310, the purge correction coefficient KAFEVACT is corrected so that the amount of fuel supplied to the internal combustion engine does not exceed the required fuel amount. May be. In this case, “KAFEVACT ← KAFEVACT + (KREFX−KREF) × PGRATE × QRATE” is executed. By doing so, the purge correction coefficient KAFEVACT is corrected so as to increase, and an amount of fuel suitable for the required fuel is supplied to the internal combustion engine.
[0123]
FIG. 8 is a flowchart for calculating the target purge correction coefficient KAFEVACZ, which is executed in step S301 of FIG. In step S351, the purge flow rate QRATE is obtained. This is calculated according to the equation (12) described above.
[0124]
In step S352, the purge transport delay table CPGDLYRX is accessed, and the purge transport delay CPGDLYRX is obtained based on the engine speed NE. As described above, the purge transport delay CPGDLYRX indicates a time delay from when the evaporated fuel is purged to the purge passage via the purge control valve until it reaches the intake system of the engine. In this embodiment, the purge transport delay CPGDLYRX is represented by the integer n. FIG. 10 shows an example of the purge transport delay table CPGDLYRX table. As shown in the table, the purge transport delay CPGDLYRX increases as the rotation NE increases. This is because as the engine speed NE increases, the number of intake steps in the internal combustion engine that are executed after the evaporated fuel is purged and reaches the engine intake system increases.
[0125]
In step S353, the ring buffer is shifted by one. The ring buffer is composed of, for example, 16 buffers KAFEVRT0 to KAFEVRTf. Shifting the ring buffer indicates shifting the data stored in KAFEVRT0 to KAFEVRTe to KAFEVRT1 to KAFEVRTf, respectively. The data stored in KAFEVRTf is discarded. Thus, KAFEVRT0 is emptied.
[0126]
In step S354, the purge flow rate QRATE multiplied by the duty rate PGRATE calculated in step S126 of FIG. 4 is stored in the buffer KAFEVRT0. Thus, “PGRATE × QRATE” calculated in the current cycle is stored in the zeroth buffer. First, second,. . . , Respectively (ie, KAFEVRT1, KAFEVRT2,...), Respectively,. . . “PGRATE × QRATE” calculated in this cycle is stored.
[0127]
Proceeding to step S355, the target purge correction coefficient KAFEVACZ is obtained according to the aforementioned equation (10). Specifically, the vapor concentration coefficient KAFEV is multiplied by KAFEVRTn corresponding to the purge transport delay value n obtained in step S352 to calculate the target purge correction coefficient KAFEVVACZ. For example, if the purge transport delay CPGDLYRX is “4”, the value of KAFEVRT stored in the buffer KAFEVRT4 is used.
[0128]
FIG. 9 is a flowchart for obtaining the vapor concentration coefficient KAFEV. This flowchart is repeatedly executed at regular time intervals (for example, 10 milliseconds).
[0129]
In step S401, it is determined whether air-fuel ratio feedback control is currently being executed. If the feedback control is not in progress, this routine is exited. If the feedback control is in progress, the routine proceeds to step S402. In step S402, if the purge flow rate QPGC is zero, it indicates that the evaporated fuel is not purged in the current cycle, so the routine is exited. If the purge flow rate QPGC is not zero, the process proceeds to step S403.
[0130]
In steps S403 and S404, the DKAFEVXH table and the DKAFEVXL table are accessed to obtain the high side determination value DKAFEVXH and the low side determination value DKAFEVXL based on the intake air amount QAIR. FIG. 11 shows an example of this table. The high side and low side determination values DKAFEVXH and DKAFEVXL are set to decrease as the intake air amount QAIR increases.
[0131]
If the difference between the air-fuel ratio learning value KREFX and the low-side determination value DKAFEXL is larger than the air-fuel ratio correction coefficient KAF in step S405 and the actual air-fuel ratio coefficient KACT is larger than the target air-fuel ratio coefficient KCMD in step S406, the influence of the purge It shows that the air-fuel ratio is in a rich state with respect to the target air-fuel ratio. Therefore, the vapor concentration coefficient KAFEV is increased by a predetermined value DKEVAPOP (for example, 0.05) (S407).
[0132]
On the other hand, when the air-fuel ratio learning value KREFX plus the high-side determination value DKAFEVXH is smaller than the air-fuel ratio correction coefficient KAF (S408) and the actual air-fuel ratio coefficient KACT is smaller than the target air-fuel ratio coefficient KCMD (S409), Indicates that the fuel ratio is lean with respect to the target air fuel ratio. Therefore, the vapor concentration coefficient KAFEV is reduced by a predetermined value DKEVAPOM (for example, 0.08) (S410).
[0133]
When the determination steps of both steps S405 and S408 are No, it indicates that the air-fuel ratio correction coefficient KAF is between the high side determination value DKAFEVXH and the low side determination value DKAFEVXL. In this case, in steps S411 to S414, the vapor concentration coefficient KAFEV is obtained according to the difference between the learned values KREF and KREFX. If the learned value KREF is smaller than the learned value KREFX, the process proceeds to step S412. If the learned value KREF is greater than or equal to the learned value KREFX, the process proceeds to step S414.
[0134]
In step S412, if the air-fuel ratio correction coefficient KAF is smaller than the learned value KREFX, it indicates that the air-fuel ratio is inclined to the rich side due to the influence of the purge. The vapor concentration coefficient KAFEV is updated by adding a value obtained by multiplying the learning value KREFX by subtracting KREF and a predetermined value CAFEV (for example, 0.02). In this case, since KREF <KREFX, the vapor concentration coefficient KAFEV is increased.
[0135]
On the other hand, if the air-fuel ratio correction coefficient KAF is larger than the learned value KREFX in step S414, it indicates that the air-fuel ratio is leaning toward the lean side. The vapor concentration coefficient KAFEV is updated by adding a value obtained by multiplying the learning value KREFX by subtracting KREF and a predetermined value. In this case, since KREF> KREFX, the vapor concentration coefficient KAFEV becomes small.
[0136]
The preferred embodiments of the present invention have been described above. The present invention can be applied not only to a mode in which a purge control valve is opened to perform a leak determination to depressurize the evaporated fuel discharge suppression system but also to other modes in which the system is depressurized for other purposes Can do.
[0137]
【The invention's effect】
According to the present invention, even when a large amount of evaporated fuel is purged by opening the purge control valve greatly in order to perform the leak determination for the evaporated fuel discharge suppression system, appropriate correction for the fuel injection amount is quickly performed. Therefore, an appropriate air-fuel ratio can be realized.
[Brief description of the drawings]
FIG. 1 is a diagram schematically showing an internal combustion engine and a control device thereof according to one embodiment of the present invention.
FIG. 2 is a functional block diagram of a control device for an internal combustion engine according to one embodiment of the present invention.
FIG. 3 is a flowchart for calculating an intake air amount QAIR according to one embodiment of the present invention.
FIG. 4 is a flowchart for calculating a duty PGCMD for driving a purge control valve according to one embodiment of the present invention.
FIG. 5 is a flowchart for calculating a purge flow rate QPGC according to one embodiment of the present invention.
FIG. 6 is a flowchart for calculating an air-fuel ratio correction coefficient KAF according to one embodiment of the present invention.
FIG. 7 is a flowchart for calculating a purge correction coefficient KAFEVACT according to one embodiment of the present invention.
FIG. 8 is a flowchart for calculating a target purge correction coefficient KAFEVACZ according to one embodiment of the present invention.
FIG. 9 is a flowchart for calculating a vapor concentration coefficient KAFEV according to one embodiment of the present invention.
FIG. 10 is a diagram showing an example of a CPGDLYRX table for obtaining a purge transport delay based on an engine speed NE according to one embodiment of the present invention.
FIG. 11 is a diagram showing an example of a table for obtaining high-side and low-side determination values DKAFEVXH and DKAFEVXL of the air-fuel ratio learning value based on the intake air amount QAIR according to one embodiment of the present invention.
[Explanation of symbols]
1 Engine 2 Intake pipe
5 ECU 6 Fuel injection valve
9 Fuel tank 27 Purge passage
30 Purge control valve

Claims (1)

A control device for an internal combustion engine having a purge system for purging evaporated fuel generated in a fuel tank to an intake system via a purge control valve,
A fuel amount calculating means for calculating a required fuel amount to be supplied to the internal combustion engine based on an operating state of the internal combustion engine;
Calculation means for calculating a value representing the amount of evaporative fuel that will be purged into the intake system by the purge system,
From the time of starting the purge, and increasing means for gradually increasing the purge correction amount for correcting the required fuel amount from a predetermined initial value to the value calculated by the calculation means,
Purge correction means for correcting the required fuel amount by subtracting the purge correction amount gradually increased by the gradually increasing means from the required fuel amount;
A judgment means for judging whether or not an abnormality judgment for judging whether or not there is an abnormality in the purge system is performed based on a pressure change when the purge control valve is opened and the purge system is set to a negative pressure ;
If it is determined by the determination means that the abnormality determination has been performed, the purge correction means prohibits the gradual increase of the purge correction amount by the gradual increase means and sets the value calculated by the calculation means to the purge Correcting the required fuel amount by subtracting from the required fuel amount as a correction amount ;
Control device for internal combustion engine.

Images